System and method for wavelet-based gait classification

• US 8,548,740 B2
• Filed: 10/07/2010
• Issued: 10/01/2013
• Est. Priority Date: 10/07/2010
• Status: Expired due to Fees

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Hesch, Joel, Ma, Yunqian, Lommel, Peter
• Primary Examiner(s): Tarcza, Thomas
• Assistant Examiner(s): ALHARBI, ADAM MOHAMED
• Application Number:
  US12/900,315
• Publication Number:
  US 20120089330A1
• Time in Patent Office: 1,090 Days
• Field of Search: None
• US Class Current: 701/479
• CPC Class Codes: A61B 5/112 Gait analysis
• CPC Class Codes: A61B 5/725 using specific filters therefor, e.g. Kalman or adaptive filters A61B5/316 take precedence
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: G01C 21/16 by integrating acceleration or speed, i.e. inertial navigation
• CPC Class Codes: G01C 22/006 Pedometers

Abstract

A motion classification system comprises an inertial measurement unit configured to sense motion of a user and to output one or more channels of inertial motion data corresponding to the sensed motion; and a processing unit configured to calculate a coefficient vector for each of the one or more channels based on a wavelet transformation of the respective inertial motion data, and to select one of a plurality of gaits as the user'"'"'s gait based on the calculated coefficient vector of at least one of the one or more channels and on a plurality of templates, each template corresponding to one of the plurality of gaits.

Claims

1. A motion classification system comprising:

an inertial measurement unit configured to sense motion of a user and to output a plurality of channels of inertial motion data corresponding to the sensed motion; and

a processing unit configured to calculate a coefficient vector for each of the plurality of channels based on a wavelet transformation of the respective inertial motion data, and to select one of a plurality of gaits as the user'"'"'s gait based on the calculated coefficient vector of at least one of the plurality of channels and on a plurality of templates, each template corresponding to one of the plurality of gaits;

wherein each template is associated with a respective subset of the plurality of channels based on the respective gait to which each template corresponds;

wherein the processing unit is configured to compare each template to the coefficient vector of each channel in the respective subset of the plurality of channels.

2. The motion classification system of claim 1, wherein the processing unit is further configured to calculate a distance-traveled estimate based, in part, on the selected gait;

wherein the processing unit is configured to update a navigation solution based on the distance-traveled estimate.

3. The motion classification system of claim 2, wherein the motion classification system further comprises:

one or more aiding sensors each configured to sense motion of the user and output signals corresponding to the user'"'"'s motion;

wherein the processing unit is further configured to update the navigation solution based on the signals from the one or more aiding sensors.

4. The motion classification system of claim 1, wherein, during a training phase, the inertial measurement unit is configured to output a plurality of channels of inertial motion data;

and

wherein, during the training phase, the processing unit is configured to receive user input indicating actual user gait corresponding to the inertial motion data from the plurality of channels;

to segment the inertial motion data from each of the plurality of channels, and to calculate a coefficient vector for each of the plurality of channels based on a wavelet transformation of the respective segmented inertial motion data;

wherein the processing unit is further configured to correlate the coefficient vector for each of the plurality channels with the user input indicating actual user gait to generate the plurality of templates.

5. The motion classification system of claim 4, wherein, during the training phase, the processing unit is configured to correlate the coefficient vector for each channel in a subset of the plurality of channels with the user input indicating actual user gait to generate the plurality of templates.

6. The motion classification system of claim 5, wherein, during an identification phase, the inertial measurement unit is configured to output a plurality of channels of inertial motion data;

and

wherein, during the identification phase, the processing unit is configured to select one of the plurality of gaits as the user'"'"'s gait based on the coefficient vector for each channel of a subset of the plurality of channels.

7. The motion classification system of claim 4, wherein, during the training phase, the processing unit is configured to generate a single template model for each of the plurality of gaits, each template model describing a distribution of coefficient vectors corresponding to the respective gait.

8. The motion classification system of claim 7, wherein, during an identification phase, the processing unit is configured to select one of the plurality of gaits as the user'"'"'s gait based on fitting the coefficient vector for at least one of the one or more channels to the respective template model of the selected gait.

9. A method of classifying a user'"'"'s gait, the method comprising:

receiving, in a processing unit, inertial motion data from a plurality of data channels;

calculating, in a processing unit, a coefficient vector for each of the plurality of data channels based on a wavelet transform of the respective inertial motion data; and

selecting, in a processing unit, one of a plurality of gaits as the user'"'"'s gait based on the calculated coefficient vector of at least one of the plurality of data channels and on a plurality of templates, each template corresponding to one of the plurality of gaits;

wherein each template is associated with a respective subset of the plurality of data channels based on the respective gait to which each template corresponds; and

wherein selecting one of the plurality of gaits comprises comparing a respective subset of the coefficient vectors to each template, each respective subset of the coefficient vectors corresponding to the respective subset of the plurality of data channels associated with each template.

10. The method of claim 9, further comprising:

generating the plurality of templates, wherein generating the plurality of templates comprises;

receiving inertial motion data from a plurality of data channels;

receiving user input data indicating the gait mode corresponding to the inertial motion data from the plurality of data channels;

segmenting the inertial motion data for each of the plurality of data channels;

calculating a coefficient vector for each of the plurality of data channels; and

correlating the coefficient vector for each of the plurality data channels with the user input indicating actual user gait.

11. The method of claim 10, wherein segmenting the inertial motion data for each of the plurality of data channels comprises calculating an average step for each of the plurality of data channels based on the segmented inertial motion data.

12. The method of claim 10, wherein generating the plurality of templates further comprises:

selecting a subset of the plurality of data channels; and

wherein correlating the coefficient vector comprises correlating the coefficient vector for each of the data channels in the subset of the plurality of data channels with the user input indicating actual user gait.

13. The method of claim 10, wherein generating the plurality of templates further comprises:

generating a single template model for each of the plurality of gaits, each template model describing a distribution of coefficient vectors corresponding to the respective gait.

14. The method of claim 9, further comprising:

calculating a distance-traveled estimate based on the selected gait; and

updating a navigation solution based on the distance-traveled estimate.

15. A program product comprising a non-transitory processor-readable medium on which program instructions are embodied, wherein the program instructions are configured, when executed by at least one programmable processor, to cause the at least one programmable processor to:

calculate a coefficient vector for each of a plurality of data channels based on a wavelet transform of respective inertial motion data for each of the plurality of data channels; and

select one of a plurality of gaits as the user'"'"'s gait based on the calculated coefficient vector of at least one of the plurality of data channels and on a plurality of templates, each template corresponding to one of the plurality of gaits;

wherein each template is associated with a respective subset of the plurality of data channels based on the respective gait to which each template corresponds; and

wherein the program instructions are further configured to cause the at least one programmable processor to compare each template to a respective subset of the coefficient vectors, each respective subset of the coefficient vectors corresponding to the respective subset of the plurality of data channels associated with each template.

16. The program product of claim 15, wherein the program instructions are further configured to cause the at least one programmable processor to generate the plurality of templates by:

segmenting the inertial motion data for each of a plurality of data channels;

calculating a coefficient vector for each of the plurality of data channels; and

correlating the coefficient vector for each of the plurality data channels with user input indicating actual user gait.

17. The program product of claim 16, wherein the program instructions are further configured to cause the at least one programmable processor to calculate an average step for each of the plurality of data channels based on the respective segmented inertial motion data.

18. The program product of claim 16, wherein the program instructions are further configured to cause the at least one programmable processor to generate the plurality of templates by:

selecting a subset of the plurality of data channels; and

correlating the coefficient vector for each of the data channels in the subset of the plurality of data channels with the user input indicating actual user gait.

19. The program product of claim 16, wherein the program instructions are further configured to cause the at least one programmable processor to generate the plurality of templates by generating a single template model for each of the plurality of gaits, each template model describing a distribution of coefficient vectors corresponding to the respective gait.

20. The program product of claim 19, wherein the program instructions are further configured to cause the at least one programmable processor to select one of the plurality of gaits as the user'"'"'s gait by fitting the coefficient vector for the at least one of the one or more data channels to the respective template model of the selected gait.